In this Article:
Understand how an LLM works
Tips to improve prompts
Prompt Components
ROSES Framework
CARE Framework
ERA Framework
How does an LLM “think”?
Text prompts are how users interact with GPT models. As with all generative language models, GPT models attempt to produce the next series of words that are most likely to follow from the previous text. As you develop more complex prompts, it's helpful to keep this fundamental behavior in mind. Regardless of the prompt that is provided, the model is simply responding with what it determines is most likely (given its training data and training targets). If you ask a question in your prompt, the model isn't following a separate “Q&A” code path, but rather it appears to answer the question because an answer is the most likely sort of response for the given question as input. That’s where prompt engineering comes in place, as it bridges the gap between human input and LLM response.
3 quick tips to improve your prompts
Prompt Components
Breaking prompts into distinct chunks, like system messages, is crucial in prompt engineering for several reasons. Firstly, it ensures clarity by separating instructions and roles, which helps the AI understand its tasks better. The system messages define the AI’s behavior and constraints, providing a clear framework. User inputs, on the other hand, contain the actual queries or commands. The knowledge can consist of all kinds of data files such as PDFs or Excel Sheets. The LLM will use this knowledge as a foundation to find an appropriate response to the user input. This separation minimizes ambiguity, leading to more accurate responses. Moreover, it simplifies debugging and fine-tuning by allowing targeted adjustments to specific parts of the prompt. A basic prompt structure is displayed below:
Prompt Engineering Frameworks
Prompt engineering frameworks are essential tools in optimizing the interaction between users and AI models. These frameworks provide structured methodologies to design, refine, and evaluate prompts, ensuring that AI systems generate precise and relevant responses.
Below we are listing 3 Frameworks to help you write better prompts.
Roses Framework
Care Framework
ERA Framework
Ready to elevate your ServiceNow Platform?
Discover the full potential of your ServiceNow platform with our help. Prompt engineering is just the start. Our experts can guide you through advanced AI concepts like Retrieval-Augmented Generation (RAG) and fine-tuning specifically for ServiceNow. Have questions or need personalized advice? We're here to help you leverage Now Assist capabilities and support your AI strategy. Contact DT Advisory today and let’s unlock the full potential of AI for your business together.